OF ARTS AND SCIENCES. 861 



The immediate purpose of the experiment is to present to 

 the two eyes the respective component pictures, not simiil- 

 taneously, as in the usual mode of binocular combination, but 

 in succession or rapid alternation. This is done most simply 

 by placing one of the common twin drawings, as of a crystal 

 or other solid traced by white lines on a black ground, at a 

 distance of about twice the limit of distinct vision, and, by a 

 proper arrangement, moving backward and forward over the 

 face of the drawing a slip of black pasteboard, so as alternately 

 to cover and expose first one and then the other picture, tak- 

 ing care that no part of one shall be in view while the other 

 is wholly or in part revealed. If, while the moving screen is 

 briskly vibrated, the optical axes be converged to an interme- 

 diate point, as in the combination of twin pictures by what 

 has been falsely termed the squinting process, the observer will 

 see the resultant picture in the same position, and with as com- 

 plete relief, as if both pictures were permanently uncovered. 

 As might be expected, the same effect is obtained with pic- 

 tures viewed in the stereoscope ; and in this case the experi- 

 ment is most readily made with transparent slides and an 

 opaque screen, caused to vibrate or to revolve near the surface 

 of the slide on which the light is received. As in these ob- 

 servations the corresponding points of the twin drawings can 

 never be seen at the same time, it is inconceivable that any 

 adjustments of the optic axes can be made to unite them pair 

 by pair, as is claimed by the theory of Brewster. The resul- 

 tant binocular perception is here due to a present picture in 

 the one eye, combined with the picture previously made in the 

 other, and which, by the well-known law of visual sensibility, 

 continues its impression for a short time after the occlusion 

 of the light. But this latter, having been impressed on par- 

 ticular parts of the retina, cannot be shifted to other parts 

 point by point, as would be necessary to efifect the combina- 

 tion with the corresponding actual ray-impressions on the 

 other eye. This experiment, therefore, confirms the conclu- 

 sion drawn by Professor Rogers at a former meeting from 



VOL. IV. 46 



